Ayoub Charef

Mr. Ayoub Charef


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 May 2024 in JUCS - Journal of Universal Computer Science
Verified
Journal Article
Open access
PDF file
ACAyoub Charef
Ayoub Charef
ZJZahi JARIR
Zahi JARIR
MQMohamed Quafafou
Mohamed Quafafou
Traffic violations by motorcycle riders pose a significant risk to urban road safety. In this paper, we present a novel approach that utilizes computer vision algorithms to detect and quantify traffic violations committed by motorcycle riders. These violations include non-compliance with helmet regulations, illegal lane changing, wrong way driving, weaving between vehicles, and running red lights. To enhance the awareness of motorcycle riders regarding the infractions they commit and the potential hazards these pose to road safety, we have developed a mobile application. This application not only provides riders with valuable feedback but also encourages them to be more conscientious and responsible for their actions. A series of experiments was conducted in the city of Marrakech, Morocco, demonstrating the system's effectiveness in positively influencing the behavior of motorcyclists on urban roads.


17 May 2025 in Results in Engineering
Verified
Journal Article
Open access
ACAyoub Charef
Ayoub Charef
ZJZahi Jarir
Zahi Jarir



1 June 2025 in International Journal of Electrical and Computer Engineering (IJECE)
Verified
Journal Article
Open access
PDF file
ACAyoub Charef
Ayoub Charef
MRMustapha Riad
Mustapha Riad
Urban traffic systems are increasingly burdened by the rising prevalence of motorcycles, particularly in cities like Marrakech where they significantly influence traffic dynamics and congestion. This paper investigates the impact of motorcycle positioning on start-up lost time at signalized intersections, employing a comprehensive methodology that integrates real-world data collection and advanced simulation techniques. Using mobile phone cameras, traffic data were captured at key intersections, and the positioning and movements of motorcycles were analyzed using the YOLOv10 deep learning algorithm. These empirical data informed simulations carried out with the simulation of urban mobility (SUMO) tool to explore various motorcycle positioning strategies. The study reveals that motorcycles positioned close to cars exacerbate congestion, extending travel times and increasing queue lengths. Conversely, scenarios with dedicated motorcycle lanes demonstrate reduced congestion and smoother traffic flows. These findings highlight the critical role of strategic motorcycle positioning in enhancing urban traffic efficiency and suggest that dedicated motorcycle lanes could significantly improve overall traffic management.